retailer data
What Walmart Scintilla Gives Suppliers
Choose a Scintilla product for your supplier question, write a report request another analyst can reproduce, and check common comparison errors.
Walmart Scintilla is Walmart’s suite of first-party data products, previously called Walmart Luminate. Start with a specific question: weekly sales movement, shopper behavior, a fieldwork task, or an approved media decision. These can require different products and permissions.
The report request template and synthetic examples below use public product descriptions. Check your organization’s access before requesting a product, field, or feed; this guide does not cover private report screens.
Match the question to the product
Walmart’s public product overview describes Report Builder as a way to customize, schedule, and share operational and omnichannel reports. Shopper Behavior provides basket and transaction insights, including loyalty and switching. Choose the product that covers the question you need to answer.
| Question | Starting point | What to verify |
|---|---|---|
| Where did weekly sales change? | An authorized operational report, such as Report Builder | Measures, channels, item/store scope, reporting dates, and refresh state |
| How are shoppers buying across the basket? | Shopper Behavior | Population, transaction scope, comparison, and definitions |
| What should a field representative do in a store? | Scintilla In-Store | Assigned task, item/store context, and field access |
| What data can an agency use for media decisions? | Approved Media Data Feed access | Supplier authorization, permitted elements, recipient, and use |
Scintilla In-Store supports third-party fieldwork. Media Data Feed supports supplier-controlled sharing with approved partners. Neither description means every supplier or agency has the same access.
Write a request another analyst can reproduce
“Send me the Scintilla sales report” leaves too many choices unresolved. Specify the request before someone exports the file.
Download the report request template (.md). Copy it into your team’s working notes. This is TrueShelf’s suggested process, not a Walmart-required form.
State the decision and the authorized source. Define what each row represents, the exact current and comparison dates, and the included items, stores, and channels. Include the measure definition and unit, refresh time, and independent control totals. Finish with the evidence needed, an owner, and a due date.
For instance, a request to explain one item’s decline should identify that item, the two complete periods, included stores and channels, returns treatment, and the store-level detail needed to reproduce the total. Record the actual report or dataset name available to your organization. A familiar metric label alone does not establish a source definition.
Complete the template inside your approved workspace. It does not require sharing customer records or credentials with TrueShelf.
Example 1: a total masks a channel shift
These synthetic numbers use generic channel labels. The labels do not describe a Scintilla export schema.
| Channel | Prior complete week | Current complete week | Change |
|---|---|---|---|
| In-store | $8,000 | $7,000 | −$1,000 |
| Pickup | $2,000 | $3,000 | +$1,000 |
| Combined | $10,000 | $10,000 | $0 |
The combined amount is unchanged, although in-store sales fell 12.5%. Comparing current in-store sales of $7,000 to the prior combined $10,000 would produce a misleading 30% decline.
Check that channels have comparable definitions and that transactions are not counted twice. With those checks complete, you can report the shift in channel mix. You still need evidence to explain why shoppers changed channel.
Example 2: sales and inventory have different time behavior
Suppose a synthetic item sells 10 units on each of seven days, with an end-of-day inventory balance of 100 units each day. Weekly sales sum to 70 units. Adding the inventory observations produces 700 unit-days of observations, not an end-of-week balance of 700 units. The final balance is 100 units; the simple average of the daily observations is also 100.
Confirm whether your inventory measure is a point-in-time balance, period average, or another calculation. Do not infer its aggregation from its label. A Friday balance also cannot establish whether the item was available every hour of the week.
Example 3: an incomplete week creates a false decline
A seven-day prior week with $100 of sales per day totals $700. A current extract through day six totals $600 at the same daily rate. Reporting a 14.3% decline would mistake a missing day for changed performance.
Record the last complete date before comparing. Use the Walmart fiscal calendar for week boundaries and document intentional partial-period comparisons.
From the request to the decision
Reconcile detail to the source total, then rank the item and store dollar changes. Use supporting data from the same scope and period to investigate those changes, and record what remains unknown. The weekly sales worksheet provides a completed three-item example with formulas and control totals.
Source definitions and access can change. Check your current agreement and Walmart documentation before using a field or sharing a result. This page does not claim Walmart endorsement or replace supplier-specific instructions.
Sources
- Walmart Data Ventures product overview — Walmart Data Ventures
- Walmart Data Ventures Announces New Name for Insights Ecosystem — Walmart Inc. ·
- Introducing Scintilla In-Store — Walmart Inc. ·
- Introducing Scintilla Media Data Feed — Walmart Connect ·